op7418/guizang-material-illustration — explained in plain English
Analysis updated 2026-05-18
Turn an abstract concept, process, or system relationship from an article into a labeled explanatory diagram.
Redraw a rough chart screenshot or raw dataset as a cleaner material style chart with accurate data.
Generate a reference researched illustration for an unfamiliar brand, scientific instrument, or historical object.
| op7418/guizang-material-illustration | angular/web-codegen-scorer | ossu/code-of-conduct | |
|---|---|---|---|
| Stars | 746 | 744 | 752 |
| Language | — | TypeScript | — |
| Last pushed | — | — | 2021-01-12 |
| Maintenance | — | — | Dormant |
| Setup difficulty | easy | moderate | easy |
| Complexity | 2/5 | 3/5 | 1/5 |
| Audience | vibe coder | developer | general |
Figures from each repo's GitHub metadata at analysis time.
Installs as a skill via npx skills add or by cloning the repo into the local Claude Code skills folder.
guizang-material-illustration is a skill for AI coding agents like Claude Code and Codex that turns articles, notes, chart screenshots, product ideas, work reports, teaching material, and cultural or historical topics into a single central illustration with short Chinese text labels baked directly into the image, in a soft 3D material art style. It aims to solve the common problem of needing one clear central picture inside a social media card, slide, article, or document that actually explains something, rather than a pretty but meaningless decoration. The skill focuses on three kinds of output: explanatory diagrams that turn abstract concepts, processes, or system relationships into labeled pictures, chart beautification that reads the real data and structure out of a rough screenshot or raw numbers and redraws it as a cleaner material style chart, and reference assisted illustration that first looks up facts and visual cues for unfamiliar concepts, brands, scientific instruments, or historical objects before drawing them in the same consistent style. It is a companion project to a separate social card skill by the same author, which handles the title, body text, theme color, and overall card layout, while this skill only produces the central image that goes inside that card. It installs with a single command using the skills command line installer, or by asking a shell capable AI agent to clone the repository into the local Claude skills folder and confirm that the main skill file and its supporting folders exist. Once installed, a user simply asks in plain language for a labeled explainer image, a redrawn chart, or a diagram for a specific concept, and the skill automatically decides internally whether the result should be a flowchart, a mechanism diagram, a chart, or an educational or cultural illustration rather than asking the user to pick a mode. It always keeps chart data and axis values accurate rather than inventing them, and it checks Chinese text labels, data, and cropping before finishing. The project explicitly states what it is not meant for, including full social card layout, complete slide deck design, real photo retouching, portrait photography, or long poster layouts, all of which belong to other tools. Its roadmap includes adding a browsable gallery of example outputs and support for more chart types like radar charts and timelines.
An AI agent skill that turns articles, charts, and concepts into a single labeled explanatory illustration in a soft 3D material style.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
double-check against the repo, no cap.